Automata networks and methods for obtaining optimized dynamically reconfigurable computational architectures and controls

• US 5,602,964 A
• Filed: 05/21/1993
• Issued: 02/11/1997
• Est. Priority Date: 05/21/1993
• Status: Expired due to Fees

• Current Assignee: Barrett Holding LLC
• Original Assignee: Autometric Incorporated (The Boeing Co.)
• Inventors: Barrett, Terence W.
• Primary Examiner(s): Sheikh, Ayaz R.
• Assistant Examiner(s): Dorvil, Richemond
• Application Number:
  US08/065,159
• Time in Patent Office: 1,362 Days
• Field of Search: 395/24, 395/23
• US Class Current: 706/25
• CPC Class Codes: G05B 13/027 using neural networks only
• CPC Class Codes: G06F 11/2035 without idle spare hardware
• CPC Class Codes: G06F 11/34 Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
• CPC Class Codes: G06F 18/40 Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
• CPC Class Codes: G06N 3/04 Architecture, e.g. interconnection topology

Abstract

A system for obtaining optimum performance and optimum graceful degradation from Lie algebra descriptions of a spectrum of reconfigurable network architectures, including, neural nets and cellular automata comprised of interconnected nodes. The dynamic performance of the computational process is monitored by continued extraction of Liapounov exponent indicators, reconfiguring said reconfigurable network architecture when said indicators predict non-optimum performance. The reconfigurable networks are reconfigured and compensatory adjustments are made of signal sampling performance and operating system performance of said reconfigurable network architecture, and the operating system architecture is optimized to the computational task by reconfiguration of nodal capabilities and degree of interconnectedness between nodes to obtain any Lie algebra description architectural form between ideal neural net with maximum interconnectedness and ideal cellular automata with maximum nodal capability.

Claims

1. A method for obtaining optimum performance and optimum degradation from Lie algebra descriptions of a spectrum of reconfigurable network architectures, including, neural nets and cellular automata comprised of interconnected nodes having dynamic performance characteristics of a computational process, comprising:

monitoring the dynamic performance of the computational process by continued extraction of Liapounov exponent indicators, and

reconfiguring said reconfigurable network architecture when said Liapounov exponent indicators predict non-optimum performance.

2. The method defined in claim 1, comprising making compensatory adjustments of signal sampling performance and operating system performance of said reconfigurable network architecture.

3. The method defined in claim 1 or claim 2, comprising:

optimizing said reconfigurable network architecture to the computational process by reconfiguration of nodal capabilities and degree of interconnectedness between nodes to obtain a Lie algebra description architectural form between ideal neural net with maximum interconnectedness and ideal cellular automata with maximum nodal capability.

4. A method for obtaining optimum performance and optimum degradation from Lie algebra descriptions of a spectrum of reconfigurable network architectures, including, neural nets and cellular automata comprised of interconnected nodes having dynamic performance characteristics of a computational process, comprising:

monitoring the dynamic performance characteristic of the computational process by continued extraction of Liapounov exponent indicators,

reconfiguring said reconfigurable network architecture when said Liapounov exponent indicators predict non-optimum performance,

making compensatory adjustments of signal sampling performance and operating system performance of said reconfigurable network architecture, and

optimizing said reconfigurable network architecture to the computational process by reconfiguration of nodal capabilities and degree of interconnectedness between nodes to obtain a Lie algebra description architectural form between ideal neural net with maximum interconnectedness and ideal cellular automata with maximum nodal capability.

5. A system for obtaining optimum performance and optimum degradation from a Lie algebra descriptions of a spectrum of reconfigurable network architectures, including, neural nets and cellular automata comprised of interconnected nodes having dynamic performance characteristics of a computational process, comprising:

means for monitoring the dynamic performance of the computational process by continued extraction of Liapounov exponent indicators, and

means for reconfiguring said reconfigurable network architecture when said Liapounov exponent indicators predict non-optimum performance.

6. The system defined in claim 5 comprising means for making compensatory adjustments of the operating system performance of said reconfigurable network architecture.

7. The system defined in claim 5 or claim 6, comprising:

means for optimizing said reconfigurable network architecture to the computational process by reconfiguration of nodal capabilities and degree of interconnectedness between nodes to obtain a Lie algebra description architectural form between ideal neural net with maximum interconnectedness and ideal cellular automata with maximum nodal capability.

8. A system for obtaining optimum performance and optimum degradation from Lie algebra description of a spectrum of reconfigurable network architectures, including, neural nets and cellular automata comprised of interconnected nodes having dynamic performance characteristics of a computational process, comprising:

means for monitoring the dynamic performance of the computational process by continued extraction of Liapounov exponent indicators,

means for reconfiguring said reconfigurable network architecture when said Liapounov indicators predict non-optimum performance,

means making compensatory adjustments of operating system performance of said reconfigurable network architecture, and

means for optimizing said reconfigurable network architecture to the computational task by reconfiguration of nodal capabilities and degree of interconnectedness between nodes to obtain a Lie algebra description architectural form between ideal neural net with maximum interconnectedness and ideal cellular automata with maximum nodal capability.